Development of Non-Symbolic Probabilistic Judgment in Children: Intuitive Biases and Individual Differences
Description
Research Hypothesis The study hypothesized that children's performance in non-symbolic probabilistic judgment tasks would improve with age, but intuitive biases would persist, especially in incongruent tasks where more targets lead to lower probability. Data and Methodology Participants The study involved 415 elementary school students from grades 2 to 5 in Wuhan, China, with ages ranging from 7.37 to 10.34 years. Probability Judgment Task Participants completed a paper-and-pencil task with 24 trials (12 congruent and 12 incongruent) involving choosing the box with the highest probability of drawing a target ball. The order of trials was pseudo-randomized. Key Findings Accuracy and Intuitive Biases Children performed better on congruent trials than incongruent trials across all grades. Intuitive biases did not decrease with age; in fact, fifth graders showed the largest bias. A significant interaction between grade level and task type was observed, but overall accuracy did not improve with age. Latent Class Analysis (LCA) LCA identified four subgroups: High Probability Judgment Ability (13.49%): High accuracy, low biases. Moderate Ability, Easily Influenced by Intuition (52.53%): Moderate accuracy with significant intuitive biases. Low Ability, High Reversal Bias (10.60%): Low accuracy with pronounced reversal biases. Low Ability, Easily Influenced by Intuition (23.37%): High accuracy on congruent trials, but significant intuitive biases on incongruent trials. Interpretation The findings challenge the assumption that probabilistic reasoning improves automatically with age. Instead, intuitive biases persist and may increase, highlighting the need for targeted educational interventions to address these biases. Practical Implications The study suggests that educators should design curricula that emphasize proportional reasoning and strategies to overcome heuristic thinking. Future research should explore longitudinal designs and cross-cultural comparisons to validate these findings and uncover underlying cognitive mechanisms.
Files
Steps to reproduce
Data Collection Methods and Protocols 1. Participants Recruitment Sample: The study recruited 415 elementary school students from grades 2 to 5 in Wuhan, China. Informed Consent: Written consent was obtained from the students, their parents, and their homeroom teachers before participation. 2. Probability Judgment Task Task Design: The task was adapted from previous research (Supply et al., 2020) and involved a paper-and-pencil format. Trial Structure: The task included 24 trials divided into two conditions: Congruent Trials (12 trials): More target objects (e.g., black balls) corresponded to higher probability. Incongruent Trials (12 trials): More target objects corresponded to lower probability. Filler Trials: Five filler trials were included to mask the task's purpose. Randomization: The order of congruent and incongruent trials was pseudo-randomized to avoid more than two consecutive trials of the same condition. 3. Administration Protocol Setting: The task was administered individually in the classroom setting. Duration: Each session took approximately 15 minutes. Instructions: Participants were asked to choose the box with the highest probability of drawing a target ball (black ball) based on the visual presentation of the boxes and their contents. Scoring: Accuracy was recorded for each trial, with scores reflecting the proportion of correct choices in each condition.